Funding | How to Cite
Return to Data Center
Introduction pdf   text
Variables pdf   text
Appendix pdf   text
Pre-election Respondent Booklet
Post-election Respondent Booklet
Planning Committee Report pdf
  bulletType of study:   ANES Time Series Study
bulletSample universe:   U.S. eligible voters
bulletSample composition:   cross-section, all fresh cases
bulletSample selection:   area probability
bulletNumber of waves:   2 (pre-election and post-election)
bulletCompletions:   1,211 pre-election; 1,066 post-election
bulletModes used:   face-to-face
bulletInstrument format:   laptop CAPI
bulletWeights:   V040101 , V040102
bulletAdditional files:   contextual file
bulletAssociated studies or derivative files:   none

In addition to content on electoral participation, voting behavior, and public opinion, the ANES 2004 Time Series Study contained questions from other areas of interest such as media exposure, cognitive style, and values and predispositions. Special-interest content provided significant coverage of foreign policy, including the war on terrorism and the war in Iraq. In addition, the study carried expanded instrumentation on unemployment and inflation, gender politics, and gay and lesbian politics. An exceptional feature was the insertion of the comprehensive module on representation and accountability, Module 2, from the Comparative Study of Electoral Systems (CSES), at the end of the Post-election interview.

The design of the 2004 Time Series study was marked by a return in the Time Series to the traditional combination of area probability sample, face-to-face interviewing, and all-fresh cross-section. A notable design feature was the continuation of the experiment begun in the ANES 2002 Time Series Study on the measurement of voter turnout. Other design features included a question order experiment for evaluation of the general direction of the country, the addition of followups on the importance of issues for the respondent placements on various issues, and the administration of parallel 'linked fate' questions to group member respondents who were female, black, or Hispanic.

Pre-election interviews 70 minutes in length were conducted September 7 through November 1, 2004. No interviewing was conducted on Election Day, November 2. Post-election interviews 65 minutes in length were administered November 3 through December 20, 2004. Randomization, employed for selection of half-samples to reduce overall interview length and for question order within batteries, was implemented by the CAI instrumentation. The total sample included 1,833 eligible persons (cases of unknown eligibility counted as eligible) and produced 1,211 interviews. In the post-election survey, 1,066 persons granted re-interviews.

Weight variables V040101 (pre) and V040102 (post) are available for use as full-sample weights that are the product of the household non-response adjustment factor, the within-household selection weight, and a post-stratification adjustment factor. These variables represent the same type of weighting first provided in weights created in 1997 for the 1994 and 1996 Time Series studies. This weighting post-stratifies to match the Current Population Study (CPS) estimate of the distribution of age group by education level. Weight variables using this construction are available for Time Series studies 1994-2004. (A corresponding weight is not available for Time Series studies preceding 1994 and has not been used since 2004.)

The ANES 2004 Time Series and Panel Contextual File auxiliary file was created to provide a core of information for analysts interested in examining, or gather data relating to, the 2004 general elections in a larger framework: candidate biographical data, past elections, expenditures, House and Senate member records and ratings, and district and state descriptions. The 436 records represent all U.S. Congressional districts (and, for population description, the District of Columbia) and thus may be used with both the ANES 2004 Time Series Study and the ANES 2004 Panel Study. This file is downloadable from the ANES Data Center.

No other studies or derivative files are associated with this study. (An ANES derivative file is a file which merges together related ANES studies.)